NJVC's CTO Blog

Friday, October 30, 2009

Having an agile day

Image by Kerry Buckley via Flickr

First I love the agile joke that goes like this,
A pig and a chicken are walking down a road. The chicken looks at the pig and says, "Hey, why don't we open a restaurant?" The pig looks back at the chicken and says, "Good idea, what do you want to call it?" The chicken thinks about it and says, "Why don't we call it 'Ham and Eggs'?" "I don't think so," says the pig, "I'd be committed, but you'd only be involved."

Back in ~2002 I got turned on to extreme programming (XP) development that is one of the agile methodologies. I first used it on a program that was having issues delivering code on time, with low defect and had a large back log of change requests (CRs)/ discrepancy report (DRs)... Initially it was the idea of pair programming that looked like it would help to improve code quality, but it was the overall framework that paid dividends. Both the development team, management and the customer got into a good dialogue that led to better estimating of capabilities between each release.   Long story short it really helped.

I have been around XP, SCRUM and Feature Driven Design (FDD) and like aspects of all of them. They are not a panacea nor is using them an excuse for sloppy documentation, the opposite is the case. To do agile well requires a discipline team.

Agile is based on a manifesto for software development http://agilemanifesto.org/.

I like bringing the principles into my day. I have marked up some of the principles to show how universal they are :

•    Customer satisfaction by rapid, continuous delivery of useful ~~software~~ ill in the blank>
•    Work~~ing software~~ is delivered frequently (weeks rather than months)
•    ~~Working software is the principal measure of progress~~
•    Even late changes in requirements are welcomed
•    Close, daily cooperation between Customers’s ~~business people~~ and Provider’s ~~developers~~
•    Face-to-face conversation is the best form of communication ~~(co-location)~~
•    Projects are built around motivated individuals, who should be trusted
•    Continuous attention to ~~technical~~ excellence ~~and good design~~
•    Simplicity
•    Self-organizing teams
•    Regular adaptation to changing circumstances

For fun the next time you are in a meeting ask your self who are the pigs and chickens? Said another way, how much skin does each person in the meeting have in the game. The litmus test is pigs have their bacon on the line. There is a place for both and one is not better than the other, but knowing who is who is helpful. What is the perspective of topic (e.g., User Story)? Is it from the customer’s view or the providers? If there are a number of problems that need to be racked and stacked? Use the point system of the Fibonacci sequence to impress your friends and rank problems. The sequence is 1, 2, 3, 5, 8, 13, 21... 1 being easy and 21 being really hard.

Most important in my book, at the end what usable product that is being delivered?

I hope that you have an agile day to.

Thursday, October 22, 2009

GEOINT 2009 Reflections

Another GeoInt has passed. As I sit on a United CRJ 700 aimed East, I leave San Antonio very upbeat with the knowledge that NJVC and its partners (old and new) had the opportunity to talk, raise a glass together, and break bread about new and novel ways we will work together to provide adaptive, agile, and cost effective solutions.

My assessment is that NJVC had a great event and I would like to give thanks to all the NJVC team that helped to make it happen.

One evening I was talking to Matt Langan who creates got geoint for USGIF and he was explaining to me about his Friday’s Food for Thought blogs. It’s a great idea and I told him that I would use his idea in my next blog. This is where I’ll step out my typical blog and I will write about my experience versus my normal technology focused blogs.

I like GEOINT. When I step back and ask why from a macro lens view it is a convergence of participants who care about GeoInt. Over the week it is a happening focused on a great mission. Hats off to Keith Masback and his team for putting on another great event. This year I had the opportunity to talk with many intelligent dedicated professionals that care about GeoInt and how it enables their customer’s to address their mission and business challenges. It is energizing when you meet a person who has a passion for their solutions.

My comment on the sessions is that they reflect a point in time and if I were on a trail I would see them as hash marks for navigation. A white hash mark on a tree is just as important as blue hash mark on a tree. The difference is how to use them to get to your destination.

I had the good fortune to meet Skunk Baxter and it was great to meet someone who had a balanced perspective of how fortunate we are to be Americans. For those that are born here it is a winning lottery ticket. It’s cool to know that I listened to his records as a kid and he works in my profession. You can be what ever you want to be and that is what makes America great and I hope he achieves his goal of being a poet warrior.

It would go without saying that all the events were great, but I am biased to say that the NJVC party at the Cowboy's Dancehall rocked with professional bull riders (PBR) gave an awesome show.

Playing at the Allder Golf Classic event is a worthy cause and the weather was perfect.

While I was at the BAE’s party at Howl at the Moon I tweeted the event. Jerry Rhoads who could not attend tweeted back requesting we have the dueling piano guys play Don McLean’s American Pie. Jerry it was a big hit and great suggestion. The power of technology when used for good continues to delight me.

What strike me comical is how many people I met in San Antonio whose offices are within 5 miles of mine, but it takes a once a year trek to meet. I am hopeful that as USGIF's got geoint evolves we as a community can leverage social networks and get together based on self-interests and mission problems locally and virtually versus popping on a plane. I like the idea that GeoInt event happens all year versus once a year. I am taking a personal action to get better at reaching out.

If you were there and we did not get a chance to meet or see NJVC's cloud demo feel free to contact me. Heck even if you weren’t there the offer is open.

Monday, September 28, 2009

Corporate IT Needs to be diligent

TKS Banking Solutions e-Portfolio bug in the U.S.

Image via Wikipedia

I had the pleasure of running into Albert Lulushi who is an old friend and a CIO at one of L3’s divisions. We spent time catching up on old times as well as sharing our challenges past and present when it comes to providing IT services to our companies. It appears that some of our problems are universal. Issues like Blackberry phone bills, users wanting admin rights to their desktops, providing a solution to remote users...

At NJVC we follow a service delivery model just like what we provide to our customers scaled to support our budget and polices. Our IT team does a great job of keeping our company secure; lowering costs each year and improving service.

Companies richest resources are their people who are becoming more entrenched in using pervasive computing technologies in their personal lives. These include email, mobile apps, Flickr, Facebook, Linkedin, Twitter, blogs, Wikipedia, Google, Yahoo, Bing, on-line games, etc. These technologies are redefining the way we communicate and the expectation of what is personal and work related. Add to this the digital divide. Corporate IT needs to support both digital natives and digital immigrants. There are many competent workers who have not embraced being digital natives. IT needs to support the continuum.

Corporate IT administrators face a daunting challenge to manage the usage of web 2.0 on the Internet or Intranets. The benefits of connectivity and social networks cannot be denied based on the Internet’s success. Unfortunately web 2.0 has the potential to open operations security (OPSEC) leaks. This includes exploiting; identities, creativity information, information quality, safety, data spillage, digital piracy, privacy, and information overload to name a few issues. On the open Internet it is difficult to moderate events. Standards for security are still maturing that allow for a variety of cyber attacks to be implemented with varying degrees of success that include: Cross Site Scripting (XSS), Cross Site Request Forgery (CSRF), Cross Gadget Request Forgery (CGRF), Phishing, Insufficient Anti-Automation defenses to stop an attacker from automating attacks. User needs to review the end use license agreements (EULA) and website user agreement & privacy policy to ensure they understand what they are signing up to and how this relates to their corporate technology policies.

I have never subscribed to the practice of security through obscurity. Add to this the fact that I know a range of smart people who have been victims to cyber assaults makes cyber defense an issue that has to be dealt with head on.

Training, risk management, following IT best practices and diligence are key to a good offense.

Keep safe.

References
http://www.digitalnative.org/wiki/Main_Page
http://www.secure-enterprise20.org/node/38

Tuesday, September 8, 2009

Thoughts for Countering Analytic Bias

Image via Wikipedia

Dan Doney showed me a a very good presentation on countering analytic bias. One of the points the presentation made is on the usage of recommendation engines as a mechanism to assist in overcoming bias and group think. Current technologies focused around search that include;

Key word
Tags (folksonomies)
Concepts (ontologies)
Popularity

The challenge with the list above is the notion that users are figuratively chained to the bias of what they know versus what they don’t know (e.g., novel) assuming it is relevant to their question. The challenge with relevance is that it is not static and is subject to change based on context.

While this area of focus is of great importance to improving the foundation of analytic tradecraft there is much that can be incorporated into other areas that include: service centers, management, trade space development….)

Information Retrieval and Recommendation models rely on the same 3 basic steps:

Model the information space (create index)
Model the user’s need
Rank the matches (information value estimation)

If we focus on the user’s need then the question we need to ask beyond categorization of a users role is the how to identify information that creates insight. One of the reasons that experts in a domain area are more interested in novel information than other information dimensions is because novel (fresh, new) can be a trigger to insight.

As we look to model user’s need we need to provide tools that help to differentiate between facts finding and insight.

Here is a fictitious example;

Problem:

User calls help desk because they forgot their password,
Tier 1 help desk admin: uses key word search on terms: password reset Standard Operating Procedure (SOP),
Key word search engine returns 3 pages of hits with password reset SOP at the top of the list
Tier 1 admin follows password reset SOP to finish help desk request - happy user
…

Insight

Users who have not logged onto their system in 20 or greater days have 60% probability of forgetting their password
Proposed improvement: For users with mobile devices send email with password best IT best practices email to users who have not logged onto the system in 10 days…

The simple example above shows bias can be bound by answering the question presented versus addressing root causes and being proactive to reduce the need for password resets.

The question is how do we get computers to identify insight?
What programming could have provided the same root cause analysis?

My belief is there are 2 areas that developing new solutions will help.

Visualization
Self aware data

Today users are chained to their monitor paradigm that primarily presents information in 2D. Large data sets provide information overload. There are products coming onto the market focusing on enabling spiritual machines that will allow for users to partition key word, recommendations and tags/concepts beyond the limits of the screen. I have been thinking about this for a while and hope to demo my use case in the near future so stay tuned.

The second is the maturation of self-aware data. While I was visiting Appistry and talking to their team this idea jelled in my head. One of the benefits of Web 2.0 has been the development of mashups using AJAX within the web browser. At it core AJAX is middleware for browsers. There is no denying the power that Mashups have done a great job in allowing developers to create rich web interfaces for users. With the advent of clouds I believe we will see the equivalent of mashups move to the data layer.

The use case would be something like this. As data is changed, read, updated and/or deleted (CRUD) the data will go enrich itself by key words, tags, concepts, popularity, and recommendations both in near-real time and retrospectively. When a system actor (e.g., user or computer) queries the system via a rich api the data should come back and say how about these other elements.

At the data end of the data center has much greater bandwidth and processing power. It seems plausible to me as we move to the cloud and take advantage of global file structures, distributed data bases, and parallel processing the ability to link data at the storage layer becomes easier. This will create huge metadata indexes that will need to be managed, but that is for another blog.

As we move toward making spiritual machines a reality NJVC will be working on developing innovations in both visualization and self aware data.

Sunday, August 2, 2009

ITIL and losing weight

I am very proud of the work NJVC does in ITIL and Service Management and have been thinking about blogging about it for a while. The idea for this blog came to me awhile back and I thought when I reached losing ~25 lbs would be a good time to publish. The idea behind this blog was to show how an activity could be used as a metaphor to understand why ITIL is a critical to making an operations environment more efficient.

My personal business case is to be healthy. The implementation has been to modify my diet and exercise more all at a low cost with an implementation that I can sustain. The end goal is to achieve my target weight and sustain it. In my case I can’t fire myself and get a new body so the constraint was working with my capabilities

My market analysis was to look in the mirror to validate that I needed to diet and get more exercise.

Key point: Never under estimate the power of common sense!
Key point: Asking for help from an expert will save you time and money!

An ITIL best practice is to start simple. Putting in all sorts of metrics at the beginning that I could not support is not the way to go. The key is continued improvement as you master one metric add another. The objective is to add metrics that support your business case. More is not necessarily better. Building on success is better than documenting failures. As I said I couldn’t fire myself and get another body so I need to work with what I have.

The tipping point is when an activity goes from new or novel to part of the daily routine and your monitoring progress.

Key point: There is no free lunch; it takes time management and discipline for success.

Like any business, I just could not stop what I am doing and focus all my attention on getting into shape. There are competing priorities that have to be managed and ITIL provided a framework that helps with the old adage; Practice makes perfect.

There are 5 key volumes of ITIL v3. Wikipedia or Pink Elephant are great places to learn more about ITIL or give us a call at NJVC. I have transcribed them below:

1. Service Strategy: As the center and origin point of the ITIL Service Lifecycle, the Service Strategy volume provides guidance on clarification and prioritization of service provider investments in services. More generally, Service Strategy focuses on helping IT organizations improve and develop over the long term. In both cases, Service Strategy relies largely upon a market-driven approach. Key topics covered include service value definition, business case development, service assets, market analysis, and service provider types. Processes covered include service portfolio management, demand management, and IT financial management.

2. Service Design: The ITIL Service Design volume provides good practice guidance on the design of IT services, processes, and other aspects of the service management effort. Significantly, design within ITIL is understood to encompass all elements relevant to technology service delivery, rather than focusing solely on design of the technology itself. As such, Service Design addresses how a planned service solution interacts with the larger business and technical environments, service management systems required to support the service, processes which interacts with the service, technology, and architecture required to support the service, and the supply chain required to support the planned service. Within ITIL, design work for an IT service is aggregated into a single Service Design Package (SDP). Service Design Packages, along with other information about services, are managed within the service catalog. Processes covered in this volume include service level management, availability management, capacity management, IT service continuity management, information security management, supplier management, and service catalog management.

3. Service Transition: Service transition relates to the delivery of services required by the business into live/operational use, and often encompasses the "project" side of IT rather than "BAU" (Business As Usual). This area also covers topics such as managing changes to the "BAU" environment. Topics include Service Asset and Configuration Management, Transition Planning and Support, Release and deployment management, Change Management, Knowledge Management, as well as the key roles of staff engaging in Service Transition.

4. Service Operation: Best practice for achieving the delivery of agreed levels of services both to end-users and the customers (where "customers" refer to those individuals who pay for the service and negotiate the SLAs). Service Operations is the part of the lifecycle where the services and value is actually directly delivered. Also the monitoring of problems and balance between service reliability and cost etc are considered. Topics include balancing conflicting goals (e.g. reliability versus cost etc), Event management, incident management, problem management, request fulfillment, access management, service desk. The functions include technical management, application management, operations management and Service Desk as well as, responsibilities for staff engaging in Service Operation.

5. Continual Service Improvement (CSI): Aligning and realigning IT services to changing business needs (because standstill implies decline).

The goal of Continual Service Improvement is to align and realign IT Services to changing business needs by identifying and implementing improvements to the IT services that support the Business Processes. The perspective of CSI on improvement is the business perspective of service quality, even though CSI aims to improve process effectiveness, efficiency and cost effectiveness of the IT processes through the whole lifecycle. To manage improvement, CSI should clearly define what should be controlled and measured.

CSI needs to be treated just like any other service practice. There needs to be upfront planning, training and awareness, ongoing scheduling, roles created, ownership assigned, and activities identified to be successful. CSI must be planned and scheduled as process with defined activities, inputs, outputs, roles and reporting.

This is not to say that ITIL is the only framework. There are other frameworks (e.g., COBIT, CMMI, eTOM, and ISO) that should be considered as you look to developing an environment of Service Excellence.

As NJVC’s tag line says, “Driven by your mission.”

Monday, July 27, 2009

Requirements Days 2009 - Ft. Huachuca

Image by alohateam via Flickr

At DoDIIS 2009 in Orlando Florida Major General John Custer's keynote address included an invitation to the attendees to come to hear the emerging requirements from United States Army Intelligence Center (USAIC), at Ft. Huachuca’s 2009 Requirements Days. The purpose of the conference was to provide a broad audience the opportunity to hear USAIC’s requirements and allow the attendees to provide conceptual and nontraditional solutions to USAIC in areas of interest.

Army’s objective Operations tempo is to identify new requirements; develop new capabilities; train the soldiers to operate and maintain the capability; and deliver onto the battlefield with Doctrine Organization Training Materiel Leadership Personnel (DOTLMPF) as fast as possible.

USAIC is looking for cutting edge ideas and emerging capabilities to help solve the challenges faced by the Army in an increasingly complex operational environment. The priorities and concerns are focused on enabling warfighters to efficiently dominate the battlefield.

In the words of Major General John Custer,

"You are either an agent for positive change or an insidious element of inertia!!!"

Sunday, July 19, 2009

Spiritual Machines revisited - Wearable Displays

Image via Wikipedia

In a previous blog I wrote about spiritual machines that included the following bullet point:

Eyeglasses that beam images onto the users' retinas to produce virtual reality will be developed. They will also come with speakers or headphone attachments that will complete the experience with sounds. These eyeglasses will become a new medium for advertising as advertising will be wirelessly transmitted to them as one walks by various business establishments.

As I was doing my regular technology research I came across a company called Microvision that was awarded $3.2 Million Contract by U.S. Air Force to Advance Development of Lightweight, See-Through Color Eyewear Display.

Based on the investments in this technology I have developed some use cases that I would like to explore.

An enabler for this technology is pico projection. We would like to determine the usability of pico projectors integrated with our blackberries so employees can give presentations without having to carry their laptops.
Pursue developing a use case to integrate our Blackberries with wearable displays so our Tier 1 technicians can roam around our facilities seeing staff do their work and enhance our customer service by being seen. During the day we can send trouble tickets to them – they can go to person’s office/cube or resolve the ticket remotely. I would like to see if the display would allow technicians look up, view FAQs and other service desk activities.
For staff giving presentation provide the capability for personal teleprompters
Provide customer resource management information about contacts during meetings

Hopefully we will be able to get a chance to work through these use cases and provide business value.

In the words of Yogi Berra, The difference between theory and practice is. In practice its different.

Friday, July 17, 2009

Data Center Modeling Thoughts

Image via Wikipedia

The need for more robust and greener data centers within the Federal Government continues to grow. Because of the long lead-time agencies need to make decisions about their future capacity needs with the belief they won’t exceed their existing data centers footprints capabilities today. We have passed the 10-year mark for many data centers and their power; space and HVAC requirements continue to grow without abatement. This is independent of outside forces that include growth of communities that are taxing power grids. In order for decision makers to make the best decisions possible they will need to capture and conceptualize the data in a way that clearly shows the trade space of options.

An approach to data center analysis allows the data center capabilities to be characterized and projected by four key process indicators (KPIs) shown in the bullets below. These KPIs provide foundational data for decision makers to model outcomes as they make long term and short term investment strategies in support of their data center facilities. The dimensions that comprise the KPIs include:

Modeling Systems

Compute demand
User base
Usage curves
Transaction profiles
Usage patterns
License costs

Modeling Facilities

Plant costs
Cooling costs
Energy usage
Support costs

Modeling Hardware

Energy profiles
Reliability
Capacity
Performance
Maintenance costs
Load

Modeling Human Capital

Labor costs
Skills
Training

This framework approach uses four model projections:

Analytical: Calculable and empirically verifiable through formulaic models
Structured: Uniform and machine describable structures
Workflow: Representations that describe or prescribe a process, often generate outputs (alerts, nested processes…) using procedural modeling
Conceptual: Showing relationships and inference of entity within a network model

These models can be projected against the data center’s performance, capacity, energy and personnel requirements. With a robust pedigree and lineage meta data model users can understand the impacts of changes over time. For example the assumptions made today will likely not be the same in the future. In addition the model will likely change as a greater understanding of the environment is modeled. Having a system that can show when and what changed is invaluable. This enhances testing what if scenarios driven from changing technology, business/ mission needs, funding and, environmental considerations.

Additional analytic power can be added by incorporating “beliefs or bias” to the model. This adds a more sophisticated understanding of the environment as models are projected. These dimensions could include behaviors of contracts (cost to market, cost to run and duration), regulations, policies, and desired service level objectives (SLOs) and service level agreements (SLAs).

I believe this approach can be a valuable tool in the Operations Executives toolbox.

Please contact me if you are interested in further information.

Customer Satisfaction

I was spending time with Chris Arroyo one of our key Customer Satisfaction Subject Matter Experts (SMEs) at NJVC and he motivated me to write this blog.

Measuring Customer Satisfaction is a core to who we are. Key process indicators (KPIs) are a key part of the NJVC's secret sauce for providing best of class services. Whether it’s running an enterprise service desk, media generation or providing an IT service. In this blog I will highlight the three dimensions of customer satisfaction that enable quality. They are:

• Services
• Processes
• Relationships

The goal of Services is to meet and or exceed expectations. This is the soft side of Customer Satisfaction because it is subjective to what the customer believes they should have. The challenge for both parties is communicate what the expectations are. This works best when the service provider and service consumer work as a partnership.

Quality of Processes is different from Services in that the objective here is to meet and or exceed defined service level agreements (SLAs). This is an iterative process that both parties need to work together so that over time the SLA are refined as well as the upstream and downstream business processes so that the measurements collected and analyzed are key to understanding the sensitivities of business and or mission capability.

Quality of Relationships provides the mechanism for increasing customer service awareness both internally and externally. Continuous training and development accompanied by a solid customer relationship management (CRM) system that includes a knowledge base that focuses on the customers business and mission are a foundation to accomplish this KPI. Have a culture that provides constant communications is keystone to success.

Once the three components of Customer Satisfaction are in place the keystone to success is metrics collection and using Business Intelligence (BI) to continually analyze results to look for areas of improvement.

Monday, July 13, 2009

Thoughts on Content Security

Image via Wikipedia

Edward Merrill, President and CEO, Granite Gate Corporation said, “Our enemies are not building up massive resources for an attack; they are attacking in millions of small ways every day”.   I will use his insiteful statement as a launching off point for this blog on content security? This is a complex problem. There are at least 6 fundamental building blocks that we need to solve together in a scalable and maintainable way to address content security that include,

1. A scalable Identity system
2. Standards framework that integrate software tool vendors
3. Security classification policies
4. Search engines that support multiple types of data
5. Reference implementations
6. Ease of use

For systems that support information requiring access control it goes without saying that if the information system can't identify who a user/system actor is then the system will have challenges applying an access control mechanism whether it is role or attribute based...

Users cover a range of role types and are faced with a range of information options from overabundance of data to sparse information that they need to assess and act on. Regardless of how sophisticated the system implementation is our grey matter makes assumptions about where the data/ information comes from, why it was added, who has used it, and how it has changed. This is called the Pedigree and Lineage of the data/ information. Pedigree and Lineage is crucial in order to understand the derivation of information and its state across the dimension of time. Assimilating all this information allows users discriminate the fit of data in their decision processes including all the biases human beings brings to bear. The challenge faced in implementation is how do we get systems to process like our grey matter. The current technology strategy is to use metadata and or tagging the data. Depending on when the system was built the concept of meta data and tagging abstraction may or may not have been a priority of the developers. Even if it was a priority the standards they followed run the range from proprietary to an actual standard making it difficult to normalize Pedigree and Lineage across a corpus of data.

In our ever-shrinking world, global networking is bringing us closer together in both time and diversity. Take a look at Security classification on Wikipedia to see this issue from a global perspective. Risk is a different perspective. Understanding the policies, tribal morays, public law… for disclosure and access is not a trivial activity and is being made more difficult as we enable the capability to share more information. The long pole in the tent for defining risk is ease of use. If we don’t define the risk via security and quality markings in the context of Pedigree and Lineage then you can only act on the data/ information with the precision provided.

The challenge we face is what is good enough? This is an exciting topic that I believe the keystone to solving is a solid framework for defining Pedigree and Lineage.

Tuesday, June 30, 2009

Cloud Anti-patterns

Image via Wikipedia

I had my head in the clouds this week, not in the existentialist view that would have me bring out my dog-eared copy of Jonathan Livingston Seagull by Richard Bach, but as a technology problem. I am curious that as more clouds get deployed, what will the cloud anti-patterns be?

For those that don’t follow anti-patterns. An anti-pattern is a design pattern that appears obvious but is ineffective or far from optimal in practice. In general there must be at least two key elements present to formally distinguish an actual anti-pattern from a simple bad habit, bad practice, or bad idea and it has been documented at least 3 times.

Anti-patterns have refactored solution(s) that are clearly documented, proven in actual practice and repeatable.

The first cloud anti-pattern that I’ll nominate is the Golden Hammer. The concept known as the law of the instrument, Maslow's hammer, or a golden hammer is an over-reliance on a familiar tool; as Abraham Maslow said in 1962, "When the only tool you have is a hammer, it is tempting to treat everything as if it were a nail.

I am optimistic that cloud computing will provide dynamically scalable virtualized resources as a service over the Internet and Intranet incorporating infrastructure as a service (IaaS), platform as a service (PaaS) and software as a service (SaaS)

Remember anti-patterns as you look to the clouds or find a company with experience to help you.

Monday, June 22, 2009

Technology frames context of thought

Image via Wikipedia

As we design new systems or use existing ones we should reflect on how does the information we receive from technology impact our cognitive processes versus provide information.

This thought started with my sitting and listening to my son practice piano. He was learning the Barber of Seville, but at that moment in time was more interested in hanging out with his buddies. While he was practicing I remembered the cartoon Rabbit of Seville and with a couple of key word search terms "youtube bugs bunny barber of Seville" the search engine returned the cartoon Rabbit of Seville on youtube. After watching the video together he changed his perspective on the song from practice drudgery to fun. During this event the technology results (e.g., Yahoo, MS Media player, youtube...) assisted our cognitive processes that had the effect of modifying our initial thought about the song and piano practice to fun experience that resulted in a better practice session. This could be measured in focus and quality of playing.

Lets expand this idea from an intimate encounter between two people to a global social network event happening now. I'll pick the events in Iran' election not to be political, but based scale and the amount of analysis being done on the technology surrounding this event. This is a large complex event that has global participants where the outcome is still playing out. Is technology (e.g., news, Twitter, Flickr, RSS, Facebook ...) changing our cognitive views of the events happening in the example of Iran's elections or is it just information? The speed and diversity of point of view from hundreds of cell phones tweeting, sending pictures and video appeared to create a movement. My hypothesis is the movement was caused by a shift in cognitive thought assisted by the technology.

My intent is not to contrast a positive or negative aspects of the examples, but to make the point that technology has the potential to help us have new thoughts versus information overload and can work on both small and large social network events. We should look at our application systems to understand how and when this occurs. Having said that people don't need technology to make decisions or have new thoughts and having technology does not always make for better decisions.

The heuristic pattern that I see between these two events is technology assisted in framing our emotional barometer as we received information allowing for cognitive insight to occur.

NJVC's ISO 9001-2008

On June 2, 2009 NJVC’s Media/Print Generation group received its ISO 9001-2008 certification for Quality Management. We are excited that the certification shows what we knew.

Friday, June 19, 2009

Oak Ridge National Labs Visit

I am writing this blog on my plane trip back to Washington DC after visiting Oak Ridge National Laboratories (ORNL) in Knoxville Tennessee. NJVC appreciates the time and hospitality that the ORNL staff showed us during our visit.

If you ever get a chance to visit ORNL it is hard not to be struck by the history of the birth place of the Manhattan project that led to the development of the atomic bomb. The X-10 reactor structure stands as a monument to the importance of science as a tool in democracy and the choices that policy and military leaders must make on how to use scientific advancements. The mission for nuclear nonproliferation continues to be relevant and important to the world.

ORNL today is more than just the birth place of the atomic bomb. Seeing models produced from super computers named Kraken and Jaguar for the advancement of science focusing on areas that include neutron science, renewable energy sources, nanotechnologies, biology systems and, high performance computing is a great mission. Scientific tools like the High Flux Isotope Reactor and the Spallation Neutron Source allow scientists from all over the world come to conduct basic research at these facilities that will benefit the scientific, business, and industrial communities. Applications like LandScan, sensorpedia and Verde are taking the power of supercomputer modeling and web 2.0 to solve hard problems.

The fusion of government, state, universities and private business makes for an exciting place to do research and development (R&D).

I look forward to my next trip back to ORNL and working with my new friends.

Thursday, June 11, 2009

Super Computers

Image by Pargon via Flickr

I had the opportunity to get an update from Cray the supercomputing company. The future in this area is adaptive supercomputing where the hardware can leverage the underlying technologies (e.g., scalar, vector and multi-threading). This will drive compiler and computer language companies to develop smarter compilers that will optimize software code to the hardware architecture available.

On the research front DARPA has their HPCS program that is focusing creating a new generation of high productivity computing systems. This research will hopefully create the next generation of economically viable high productivity computing systems for the national security and industrial user communities. These systems must have the following attributes:

Performance: Improve the computational efficiency and reduce the execution time of critical national security applications.
Programmability: Reduce cost and time of developing HPCS applications.
Portability: Insulate HPCS application software from system specifics.
Robustness: Improve reliability and reduce vulnerability to intentional attacks.

If the vision of HPCS is to High productivity computer systems that are optimized for users, not for benchmarks then at what point does super become common?

Wednesday, June 10, 2009

Virtual Clouds

Image via CrunchBase

I attended a seminar from Force 3 when the Nats played the Cincinatti Reds to hear about EMC, VMware and Cisco's joint venture called VCE. EMC described what they termed "private" cloud computing. On Youtube you can find a key note address during VMWorld.

VCE is bringing the capabilities that were established on mainframes (e.g., LPARS and IBM's Parallel Sysplex) to any device whether its a PC, server, blade center or mobile device. The cool part is EMC and its partners are integrating their tools like SMARTs, RSA security, replications capabilities... to provide greater control over the environment following ITIL best practices.

One of the capabilities I like is the idea of the system shutting down processors that are not being used to save on power consumption. In a large data center Green IT for sustainable development is a worthy goal. Hopefully software vendors will be able to adapt their licensing strategies so companies can take advantage of paying for what they use.

At NJVC we are constantly looking at how well we can apply new approaches to our internal Corporate IT as well as our external customers. I am a big believer in, "eating your own dog food." In a previous blog called Virtualization value we have found that virtualizing applications and storage is helping us save on hardware and support costs as well as helping us implement our business continuity plan. These same approaches are providing similar value to our customers.

The bottom line is understand your requirements and based on the size of your environment use capabilities that support them best. If you need help let us know.

Sunday, June 7, 2009

USGIF- GEOGala

I attended the USGIF GeoGala on June 5, 2009. NJVC is a proud USGIF sponsor for this event and Tech Days.

The GEOGala is a black-tie dinner event that was held at the Hyatt Regency Reston in Reston, VA. Events like this provide an opportunity to meet government and industry friends that we don't get to see during the work weeks.

One the evenings memorable events was the video by Penn State's Public Broadcasting on the GeoSpatial Revolution. Their simple point is GeoSpatial has become common and its worth reflecting on how this achievement has become such an everyday activity.

Keith Masback and his team did another great job of putting together another fun filled evening.

Monday, May 25, 2009

Observance of Memorial Day

Image by Getty Images via Daylife

As you reflect on Memorial Day remember Freedom is not free.

In observance of this day here is a prayer by By Rev. Dick Kozelka (ret)
First Congregational Church of Minnesota Minneapolis, MN.

Eternal God,
Creator of years, of centuries,
Lord of whatever is beyond time,
Maker of all species and master of all history --
How shall we speak to you
from our smallness and inconsequence?
Except that you have called us to worship you
in spirit and in truth;
You have dignified us with loves and loyalties;
You have lifted us up with your lovingkindnesses.
Therefore we are bold to come before you without groveling
[though we sometimes feel that low]
and without fear
[though we are often anxious].
We sing with spirit and pray with courage
because you have dignified us;
You have redeemed us from the aimlessness
of things' going meaninglessly well.
God, lift the hearts of those
for whom this holiday is not just diversion,
but painful memory and continued deprivation.
Bless those whose dear ones have died
needlessly, wastefully [as it seems]
in accident or misadventure.
We remember with compassion those who have died
serving their countries
in the futility of combat.
There is none of us but must come to bereavement and separation,
when all the answers we are offered
fail the question death asks of each of us.
We believe that you will provide for us
as others have been provided with the fulfillment of
"Blessed are those who mourn, for they shall be comforted."

The White House Commission on Remembrance is an independent government agency, established by Congress, whose missions include:

Promoting the spirit of unity and remembrance through observance of The National Moment of Remembrance at 3 PM local time on Memorial Day
Ensuring the nation remembers the sacrifices of America's fallen from the Revolutionary War to the present;
Recognizing those who have served and those who continue to serve our great nation and reminding all Americans of our common heritage.

Saturday, May 23, 2009

Using Clouds for fast loading of triples

Image by kasei via Flickr

The guys at BigData have run some really exciting benchmarks by loading 5 billion triples loaded in just under 10 hours for an average throughput of 135k triples per second. They say that their maximum throughput was just above 210k triples per second. 1 billion triples was reached in an astonishing 78 minutes. With these types of benchmarks the practicality of using rdf triple stores for large data problems keeps getting brighter.

Friday, May 22, 2009

Spiritual Machines

Image by Erik Charlton via Flickr

While at 2009 DoDIIS Worldwide Major General John M. Custer Commanding General, United States Army Intelligence Center and Fort Huachuca gave a thought provoking presentation at the second day's plenary session. I enjoy listening to him because his points cause me to reflect on whether or not I am providing the greatest value to the end users whether they are warfighters, policy makers or non-military users.

To make his point about what Spiritual Machines could mean to the warfighter he showed a Pattie Mae's video of technology called the Sixth Sense. The context for viewing this site was to gain insight into the concept of the Spiritual Machines. The idea being that technology will enhance our cultural and linguistic capabilities. From this video you could imagine soldiers using the technology to amplify the world around them as the sensors gather information around the soldier and help him/her process the environment better. You can view it at the TED website. If you are not a regular visitor to TED, I recommend it as I have been viewing it prior to my mention here. TED stands for Technology, Entertainment, Design. It started out (in 1984) as a conference bringing together people from those three worlds.

Since the blog is entitled Spiritual Machines I thought it made sense to me to look at the technology trends for 2010 to see what was predicted. The list below is from The Singularity Is Near: When Humans Transcend Biology that is the 2005 update of Raymond Kurzweil's 1999 book, The Age of Spiritual Machines. The list says:

Computers become smaller and increasingly integrated into everyday life.
More and more computer devices will be used as miniature web servers, and more will have their resources pooled for computation.
High-quality broadband Internet access will become available almost everywhere.
Eyeglasses that beam images onto the users' retinas to produce virtual reality will be developed. They will also come with speakers or headphone attachments that will complete the experience with sounds. These eyeglasses will become a new medium for advertising as advertising will be wirelessly transmitted to them as one walks by various business establishments.
The VR glasses will also have built-in computers featuring "virtual assistant" programs that can help the user with various daily tasks. (see Augmented Reality)
Virtual assistants would be capable of multiple functions. One useful function would be real-time language translation in which words spoken in a foreign language would be translated into text that would appear as subtitles to a user wearing the glasses.
Cell phones will be built into clothing and will be able to project sounds directly into the ears of their users.
Advertisements will utilize a new technology whereby two ultrasonic beams can be targeted to intersect at a specific point, delivering a localized sound message that only a single person can hear. This was demonstrated in the films Minority Report and Back to the Future 2.

Notice within the list a grouping around visual and auditory sensory heightening and based on today’s technology all seem reasonable. Visual and auditory sensory are one aspect. Advancements in nanotechnology that promise nanomachine delivery of medicines like insulin to diabetics or medical care to soldiers on the battlefield could make a powerful advancement to the soldier of the future.

An important question that this raises is what is the impact by exploiting the counter side of Spiritual Machines?

It is on this point I think we will discover a new field in Cyber defense called Cyber Psychology of Human Interface Design. I am a firm believer that our thoughts create our reality. I like the saying, if it wasn't for this thought I would be having a different reality. With the concern of Cyber in the world regardless of the instigators, I wonder will the next generation Cyber viruses be focused on behavior shaping? Case and point, try to pull a teenager's attention away from an interactive computer game.

As we move into the future of Spiritual Machines we need to understand both the positive and negative aspects of technology.

Saturday, May 9, 2009

DoDIIS Worldwide 2009

NJVC will be at DoDIIS Worldwide this year. As a company mantra we say, Driven by your mission so the focus of this year’s conference Empowering Decision Advantage resonates well with us. Please stop by our booth at location 1430 in the exhibit hall. I’ll be there most of the week.

As I was thinking about what I would write for this blog I came across an AFCEA article entitled, Battlefield Information Systems Empower The Warfighter by By Lt. Gen. John A. Dubia, USA Ret.) May 2002 and felt the closing paragraph says it best,

"…Despite all of this technology, there is no substitute—nor will there be a substitute—for the soldier, marine, sailor, airman or coast guardsman who is taking an objective. We must never lose sight of the fact that it is the human element—that man or woman in uniform—who is the ultimate weapon to win any war."

When we build solutions we should never lose site of that.

I look forward to seeing old friends and making new ones.

Friday, May 8, 2009

Virtualization value

Image by MysteryBee via Flickr

I am taking pause to reflect on one of the Corporate IT projects that is paying real dividends. We moved to virtualized management (VM) our environments. This has created real efficiencies for life cycle development of applications and the Corporate IT team is really jazzed about this achievement.

We now create development, maintenance and production systems very quickly that duplicate the production environments on the fly. We are leveraging the elasticity of our environment by provisioning environments when we need them and turning them off or removing them when we don't need them. For COOP and business continuity we are shipping copies of the environment around the network and don't have to worry about patching and hardware compatibility. As we try to do more with less this is one of tools in our arsenal that is helping our back office IT staff. These efficiencies are giving back time to our staff so they can focus on developing better standard operating practices and policies that at the end of the day will likely have the greatest impact on minimizing down times whether planned or unplanned. I am thankful we have such a great Corporate IT team.

Saturday, April 25, 2009

Bridging the failure gap to make IT projects successful

I have been out of the office for the last 3 weeks and as I am heading back home from Seattle to DC I wanted to blog about how to make IT projects successful. I ran into Lewis Shepherd at San Francisco’s Airport – he has a tweet about it. One of our conversation topics was about how much blogging and tweeting we do. I had said that I was going to challenge myself to write more often and here it is. I have also tried to increase my tweeting for a month. At the end of this month I’ll see if it becomes more of a tool for me.

Service and Gratitude are words that are not used enough in developing information technology (IT) systems today let alone in our daily lives. Countless articles have been written on why IT projects fail as well as case studies showing why they succeeded. Want proof click the link “Why software projects fail”. I am reading anti-patterns on my Kindle to continue my own education on this topic. It’s a fun read and I have created a few anti-patterns from my own experiences.

For any IT project we can slice and dice it from many views. From a program management (PM) lens we can analyze projects in terms of cost, schedule and resources (the triple constraint). Some PMs add the technology vector to this as well. PMs can draw cool vector diagrams that show all dimensions of earned value and technology complexity. At the technology level we can build frameworks that partition the development into definable segments that can be easily understood by the team with the goal of maximizing productivity while minimizing defect. Philosophically we can define the patterns and anti-patterns that show behaviors that have been seen at least 3 times to keep us from falling into traps. We can engage users through the whole project and alpha test, beta test, black box test, developing training plans. We can measure usage requirements changes…

Is that enough to make a project 100% successful in the customer’s eyes that are both subjective and quantifiable? (□ Yes / □ No / □ Maybe)

Hypothetically what would you be willing to sign up to in writing where the penalty is loss of fee?

For grins and allow me to make my point let’s run a model on a hypothetical IT project that I have detailed below:

You have been given a software project that needs to do . You hire a seasoned Program Management Institute (PMI) certified PM who knows earned value and brought in a top notch architect who has analyzed the patterns and anti-patterns of the project with the PM. He has built the architecture. You have a seasoned group of developers that know SCRUM with the experience of doing daily builds… Business analysts know the users business domain. Users have been assigned who are interested in supporting the project and you have adequate budget.

Will you guarantee the project be 100% successful, brought in on time and on budget with 0 defects?
Would you sign your name and reputation on the bottom line?
What do you think?

How do you fill the gap?

I assert that if 2 components are added to a project the likelihood of failure or not meeting the customer’s expectations will dramatically be minimized.

The secret sauce is:

Service
Gratitude

By the way the answer had to be simple, obvious and have nothing to do with technology. Why because we are human.

If a team truly believes they are in service to their users and have gratitude to their team mates and users (i.e., checked their egos at the door and are prepared to listen) how can any project fail assuming they have all the things above?

When I hear terms like this;

Build it and they will come
My experience has been…
Our consults know your business…
I can remember doing thing like this…
How many more can you add?

I would be suspect if I was not also hearing;

How can our team best serve you?
Our analysts will listen to you before they make recommendations of what your requirements are
This team likes to solve problems and is very inquisitive (curious) and love working together
Our team members respect each other so much that when one team member presents an idea, the others will make it better before they present their ideas
This team is resilient and approaches problems from a neutral place
How many more can you add?

You might ask if this is testable. One simple test you can try if you like is as follows.
The next time you are in conversation with someone and you are thinking of what you are going to say while the other person is talking then you’re not truly listening to them deeply.

If you are not listening then you can’t be in service or be in a state of gratitude and back to my question of the gap you have not decreased the probability of failure of an IT project.

Here is another variation of the test.
Next time someone presents an idea in a meeting or conversation. Make it better before you add your 2 cents. If you can’t then say nothing until you can. Silence is Ok. That could mean asking for clarity and allow the speaker to work more than 10 seconds on their idea. Let them mentally work to crystallize their idea. This likely means you are mostly asking clarifying questions versus applying your ideas.

If it’s a good idea it will continue if not their wisdom will cause them to move on.

When you can get a development team acting this way with their customer, you’ll be blown away by the quality of thought they produce and that will translate into a more successful project because you have bridged the failure gap. In my mind this is what a true high performance team is and not a bunch of senior grey beards that got together. There is an economy here to be appreciated. You can build multiple high performance teams this way versus there are only so many grey beards.

Let me know if you need help building your team this way.

Sunday, April 12, 2009

Mobile Device Thinking

I am visiting NJVC's staff in the Pacific. I had the good fortune to merge family spring break with work. The long flight provided me with lots of time to catch up on my reading. There were a couple of articles in Technology Review Magazine Wed Apr 1, 2009 that got me thinking. One is on an emerging technology called racetrack memory. The others were on the subject of biomedicine and their merging with computers that I will write about in future blogs.

Stuart Parkin a fellow from IBM is working on storing information in U-shaped nanowires that can theoretically store 100 times more data into the same area as a flash-chip transistor, and at a similar cost according to the article. Assuming this is true, storage is heading to small sizes that are unheard of. Another trend that I keep track of is the re-emergence of distributed files systems like hadoop. Back in the ‘90s IBM’s distributed computer environment (DCE) began to pave the way virtualized environments. Virtualized environments are becoming common place at all levels of the compute stack.

Consider if;

Bandwidth was not an issue,
Large volumes of storage can be housed in really small form factors that require low power.
Moore’s law continues for the next 10 years,
Secure distributed file systems improve
Grid and parallel computing environments continue to enhance their capabilities
Cryptography improves for applications and users
Power and cooling are still issues
Mobile devices remain popular
PC and mobile devices continue to merge (e.g., blackberry, iPhone...)
...

As I look at a Pacific Ocean sun set and think about possibilities here is something to ponder. Say in the future, 10 - 25 years out, why wouldn’t we move from our brick and mortar data centers and distribute the content across the millions of mobile devices that make up today’s and likely a larger number in the future cellular network? On the positive side it’s more distributed then any data center for availability. Consumers are constantly replacing their phones with the latest and greatest mobile devices that out paces any technical refresh cycle and is funded by the consumer. One idea for consideration is Telecommunications carriers would carve off pieces of the mobile devices processor, memory and storage for private business usage. This is my mind could be an evolution of new cloud offerings (e.g., Amazon’s EC2). At the same time with appropriate security companies would provision their employee’s desktops to their mobile device or provide one to them... I am curious what our corporate ITs would think of this? Assuming they had full access to that portion of the mobile device for moves, adds, changes, patches, deletes and, license compliance I think they would be supportive.

Let me be the first to admit that there are many obstacles to making this happen, but its fun to dream. The issues include, but are not limited to, distribution to these specially enable mobile devices, mediation, privacy, algorithms to ensure workload, security... The idea may never happen as I have written it, but the idea of utility computing where in the future data centers run on platforms that are different from the past is real. For example companies like Cisco are betting that by merging fabrics (e.g., storage & networking and compute capabilities) into single chassis that minimizes the amount of disparate hardware a data center needs and can be greener that will appeal to consumers of data center products. It's only a matter of time that someone figures out how to build a ASICs into each mobile device and they become the network versus needing our current approach to network routing and network switching. Aloha

Monday, March 2, 2009

Facebook's Thrift framework

Image via Wikipedia

It all started with the simple idea of creating a web front end to access our lab’s hadoop and hbase environment. Looking at the output from a linux cat command does not have much sizzle. I wanted an easy way to demo viewing data stored in hbase or hadoop and thought using the ruby interface would be a piece of cake. I have not done much programming in ruby so I thought this would be a chance for me to learn.

We have a 13 node hadoop cloud built for learning and to demo capabilities. We built the environment manually and learned lots of lessons learned. For example hadoop version 0.18.4 cares about compile timestamps and frowns on typos in config files. In the future we plan to use an application like puppet for managing the environment as well as ganglia for monitoring beyond our existing network operations center (NOC) capabilities.

In order to use ruby one way available to interface with hbase or hadoops api is to use facebook’s thrift. Compiling thrift was an all morning exercise of downloading missing apps from g++ compilers, ruby, automake… from the base load of the operating system (OS) we use. The sysadmin challenges of these programs are laborious and I am appreciative for all the web sites that provide answers to my questions. I did not once have to look at a traditional manual. By mid-afternoon I had the equivalent of hello world running that showed me the names of my tables in hbase.

The cool capability of thrift is I could have easily written that same app in java, python, php or C++. The goal of thrift is to provide a scalable cross-language services implementation. From the developer’s presentation:

Thrift focus is on;

Generates definitions for all the types in each language
Generates Client and Server interfaces for each language
Lets you run a .thrift file as a build script!

Thrift leaves to you;

Anything to do with sockets
Anything to do with serialization

As I have gotten older I like working in scripting languages. I guess my days of waiting ½ hour for my COBOL program to compile on the mainframe scarred me. Hadoop and hbase are a set of applictions that are letting me work out ideas of schema-less data models. An analogy that resonates with me is;

Data is trapped in schemas

I am reading Outliers by Malcolm Gladwell and one of the points he makes in his book is about being in the right place at the right time and having the opportunity to practice, practice, and more practice.

When I see programs like hadoop, hbase, thrift, Google's OpenSocial Initiative…, I see an evolution of the application frameworks based on people having a lot of practice dealing with aggregating Internet traffic and applications across the world wide web. Net net these solutions will integrate more information and make it available not just to the elite, but to the masses writ large.

Stepping back the life cycle complexity of software as a service (SaaS) will be a component of future solutions to private industry and government. The challenge is keeping up with these development innovations within the enterprise and providing life cycle development and support platforms for workers of all ages and capabilities.

My experience with thrift is what I learned over the weekend, it is a very clever approach to a hard problem. I am having fun building this environment and at this point in time will consider it within the trade space versus using a rest. Stay tuned for more updates.

Friday, October 30, 2009

Thursday, October 22, 2009

Monday, September 28, 2009

Tuesday, September 8, 2009

Sunday, August 2, 2009

Monday, July 27, 2009

Sunday, July 19, 2009

Friday, July 17, 2009

Monday, July 13, 2009

Tuesday, June 30, 2009

Monday, June 22, 2009

Friday, June 19, 2009

Thursday, June 11, 2009

Wednesday, June 10, 2009

Sunday, June 7, 2009

Monday, May 25, 2009

Saturday, May 23, 2009

Friday, May 22, 2009

Saturday, May 9, 2009

Friday, May 8, 2009

Saturday, April 25, 2009

Sunday, April 12, 2009

Monday, March 2, 2009

Subscribe to the NJVC CTO blog

CTO

Blog Archive

Share it

Links of Interest

Tag Cloud